Roast #38: V6 reflection + survey results
Let's go literally any team but the Boston Celtics
Some quick numbers:
Final sold V6 box count: 84 boxes (including test roasts) // (21.7% increase)
Delivered 121 lbs (just from coffee club boxes) // (33% increase)
57 return buyers // (68% increase)
Growth is linear. I'm just happy that growth hasn't contracted or plateaued.
Compared to V5 data, we've done an even worse job this time around at bringing in new customers.
The trend of once you try a box, it's likely that you'll stay for another box seems to hold. I hope to maintain this, as that means I'm delivering something valuable to you all — or at least valuable enough for you to give me another chance.
V6 Financials:
Try to ignore the negative profits. I couldn’t easily distinguish when the additional Wilton Benitez beans stopped contributing to profit, so I just lumped the additional costs into the Profits section.
Roasting still accounts for a large portion of my costs. I've been considering moving up to the 15kg roaster. Roasting on a larger machine has its benefits, but if the number of orders exceeds my roast batch size, then it becomes challenging to try and sell the excess.
Say each roast produces 20 lbs. If I received 25 lbs worth of orders, I'd need to do two 20-lb batches and then figure out how to sell the excess 15 lbs from the second roast.
For that reason, I'm likely going to stay on the IR-5. Unless I have an easy channel to push excess roasts to (i.e., a cafe/weekly popup where I can use the excess for service, or a lot of wholesale interest), I don't think going up yet's a great idea. Plus, I really enjoy roasting on the IR-5. <3
V6 Survey Results:
The Black Honey Sidra was the favorite this round! I would've bet the Pink Bourbon would've been the favorite.
The Castillo + Colombia was tricky, so I understand why it was not picked as the favorite. It wasn't extremely aromatic, and its profile could've been interpreted as very conventional. However, with enough rest and when drank at room temperature, I found it produced a very complex cup. It was the dark horse coffee for this round (imo).
The best individual cup though was castillo -- it was always good but one day it came out as one of the juiciest coffees I've ever had. Not sure what was different about that day but that cup was exquisite.
Black Honey Sidra - the spiced floral flavors were amazing and I found myself constantly craving it and overlooking the others
Granja El Paraiso 92 Double Anaerobic Washed Pink Bourbon. I just love how juicy it is and how you can get different flavors depending on brew temp/water. Got some guava candy at 87c and passion fruit at 92c.
Things that went well:
A lot of presales:
I was looking at my numbers mid-V6, and I got confused when I saw that I was selling more but was making less. That's when I realized that everyone was buying the discounted presale box!
Presales will be replaced with subscriptions. Having it as a "presale" gives customers more flexibility, but it keeps things risky for me as I'd always have no idea how each coffee club will perform. With a subscription, at least I have a starting number of "maybe"s.
Two months is the preferred cadence. I'll do my best to meet it, but there will be times when I miss (e.g., V7) :/.
An easy way to pause would be important for me -- get frustrated when brands only have cancel and it is convoluted to do so.
Will do. I just had a horrible experience with Instacart, where I'm locked in for a year of membership because I forgot to cancel. If you didn't intend to buy the box, I see no reason not to return your money. I'll make sure you can pause/cancel easily.
first dibs for retail bags after subscription fulfillment, some kind of perk to differentiate subscription from buying individual boxes
I like this idea. Maybe coffee club subscribers will get a discount code that they can use for additional single bags. I'll think of other fun benefits to include, but if there's an opportunity to share something one-off, I'll prioritize subscribers.
If every 1 month was an option, I would sign up for that
I love how there were actually a few of these. 🙇. I hope one day I'll be able to deliver at this cadence.
Roasting
Please keep a funky bean. And the roasts light! v6 was (in my opinion) leagues better than v5.
9.5 -- I liked V5. It was good, but it felt like a "hobby" production. This was totally different. The sourcing, the roasting, and the curation were all on a different level. This box goes toe to toe with some of the best names in roasting, and honestly I think it wins out. Kudos to you -- you really put in the work on this box, and it definitely shows.
Comments like the above mean the world to me. I'm so glad some of you found V6 an improvement over V5! I still have tons more to learn/grow, and I look forward to translating my efforts into improved future coffee clubs.
Roasting went smoothly for V6. Many of you mentioned that the roasts for V6 were a lot lighter than V5. You all are correct.
For V6, our roasts landed around 9% to 11% weight loss.
Post-roast, I cupped each batch and felt the results were acceptable. Some of you thought that the Castillo + Colombia (C + C) might've tasted a bit green. I did go pretty light with that bean as I felt the intrigue was its terroir, so I skewed towards less development to preserve the acidic characteristics of the coffee. It's possible the resulting batch could've not developed properly, so please DM me if you felt it was unfair that you had to pay for that bean, and I'd be happy to refund you for that portion.
One area where I could improve is better consistency across all roasts. Ideally, the roast curves should have little to no variance.
Here are all the pink bourbon roasts stacked on top of each other. We can see that there's still some variance between the roasts. Also the finish seems to get wonky as I’ve found myself having too much heat, and I have to drastically drop heat to ensure some development.
My curves will get better as I roast more with the IR-5 and build better processes (e.g., better between-batch protocols, better pre-roast planning, etc.).
Lastly, there was the test batch. There was little difference between how I roasted the test roast and the production roasts. The key difference I implemented was increasing development for the Pink Bourbon and the Castillo + Colombia and upping the initial charge temp for the Castillo + Colombia. When I realized I wasn't getting super green notes with an 8:40 total roast time, I felt I could push more heat initially.
Things that didn't go well:
Packaging:
The bag opening was too small
Very minor gripe, but the bags have such a narrow opening that it can be tricky sometimes to get the beans out efficiently, or put them back in...
I think the small opening of the coffee bags was a bit annoying to use in practice
If possible a wider bag rather than a narrow one. But then again, it does fit perfectly in your shipping box. And I "pour it" to weigh it. So don't listen to me(?)
There were more points concerning the bag opening that are not listed above.
I absolutely hear you. There are just more pressing things for me to focus on. I'll revisit this in the future after I've locked down my single-item (non-coffee club) packaging.
No more handwriting coffee info onto bags
Also, might be nice to have stamped or printed labels on the bags as opposed to handwritten.
Drop the hand writing (will help you to scale) and consider a package a larger opening.
No more handwriting things. Fun fact, I shortened "Castillo + Colombia" down to just "C + C" after realizing I'd have to write that on 80+ bags.
But for V7, we'll go with stamps and stickers.
Packing:
I came into V6 thinking that packing would be easy since I would use Multimodal's commercial bag packaging system.
What I should have realized was how long it took the machine to weigh out each bag dosage. Like any precision-forward instrument, the machine would slow down once it got close to the target weight. So for a target weight of 125g, the initial 120g of beans would be quick-ish, but then that last 5g would maybe take 30 seconds as the machine would drip 1-2 beans a second to ensure it didn't go over the target weight.
The above isn't to scrutinize the machine; it's a game-changer, and I love having it in my workflow. It's just that I expected that I'd be able to do all my packing (300+ bags) in less than an hour — so I had unrealistic expectations of its capabilities with only one operator (me). So, on my packing day, I scrambled when I realized I wasn't close to packing everything before they closed. I took home all the beans I couldn't pack at the facility and had to pack them with my tiny hand-operated bag sealer/mini-scale. That was not fun.
For V7, I'm going to roast and pack on different days. Separating both operations will give me more time to use Multimodal's packing equipment to ensure packing gets finished in one go.
Even larger disparity in profile preference:
More for cleaner coffees:
Definitely the "clean" funk profile and the lighter roasts. Previous boxes were either a bit too funky for me or a little toasty. This was excellent.
I like that you've gone back to unflavored beans for this round. The novelty of the previous round was awesome, but (probably just because I ordered too much) that wore off about halfway through the bags for me. Also, love how lightly roasted this round is.
About half the feedback loved the "cleaner" profiles, and the other half felt we could've pushed the funk/flavor-spectrum harder.
More for really profile-forward coffees:
7. Nothing was overly funky, and I think having at least 1 in each box would be a nice mix
5 - for some reason, the beans this time haven't tasted as flavorful as the boxes in the past
Then the other half wanted something a bit more flavorful.
I’ll try to keep a good balance. What I land with each box will really depend on how I’m feeling when sampling. At the moment, I’m leaning towards slotting in a co-ferment for V8 as it’s been a while since I’ve offered one. I was sharing the V5 peach co-ferment with friends this past weekend, and I missed how fragrant and easy it was to work with.
I’ll make sure to call out when a box is more conventional/terroir-focused, and when it’s more on the funky/process-heavy side.
Not enough focus on marketing + advertising:
We are going to have to seek out new channels for marketing (ie advertising). We used to rely on Reddit for sales, but I think I've exhausted that channel as most of my V7 posts have been removed.
I was set on trying ads for V6 but ultimately decided not to. Joshua, a past customer, offered advice (🙇) on what I should have set up before running ads. But he mentioned one point, which was "Creative and CTA will def be key," and that kind of hit in the sense that I'd be throwing away money if I were to push a post/ad that didn't have much creative thought.
I'll give advertising a go for V7.
People have also suggested figuring out a way to add video content to my flow.
Video brew guides that give a little more info about the coffee -- what you liked about it, how you started working with the producer, etc. Doesn't have to be for every bean or a large production, but it could be a nice personal touch.
Maybe some website videos of your pour over process. I ended up doing what I normally do, result was great.
Make reels or TikTok's of you during the roasting process. I think a lot of people are really curious about what the process looks like.
Some of you have suggested (offline) turning to Tim Wendelboe and some of his subscription YouTube videos for inspiration. I think there's something here. #briantheinfluencer
Closing V6 thoughts:
We need to ask for help
The saying on “if you want to go fast, go alone, if you want to go far, go together” has been kinda banging around in my head as of late.
I'm feeling that I might be holding thoughtfulcoffee back by not asking for help. A few of you have mentioned/offered, which I really appreciate! I’ll be asking for help more often in the future, so please respond if interested!
Plus, it's way more fun working alongside people than alone. I realized this recently when I did a small popup by myself, and it just wasn't as fun as working alongside with people (like Maria and Aki 🙂).
I will try to reach out for help more often.
Thank You's:
Aki + Maria. Maum Market was hectic. You both were the heroes. Thank you for being there for me and helping at that event. I would've been screwed without you both.
The helpers (Seean + Matthew + Shaka + Jake) for coming through with such short notice for the AAPI coffee roaster event. Also a big thank you to Brian Quan for considering me as a co-host for this event.
Michael from unblended for being there when I needed help. Working with you and your team has been such a pleasure. I hope you are having an amazing time in Colombia at the moment!
Everyone who has thought to showcase thoughtfulcoffee somehow:
Alex (sorry! wrong tag) for thinking of us for your Meta coffee club!
Andrei for including us in your home popup event!
Alex for having us on retail (even with our bland bags) at Paper Son Coffee
Whoever keeps bringing my bag to the coffee klatsch meet ups. I feel like I see my bag as a story in one of their events after every coffee club.
Thank you to Multimodal for always being so accommodating. From their staff being willing to share knowledge to letting us host events (cuppings), they've been an amazing roasting facility to work with.
Ethan + Anton — thank you for all the help with the color sorter earlier today. I’m excited to explore how to insert that tool in my processes!
Thanks for the write up! Great work as usual. I’m looking forward to v7.